It is widely known that a central nervous system (CNS)-neuron in higher organisms such as human is not capable of regeneration once injured. Therefore, one who has received an injury on his (her) spinal cord due to, for example, a traffic accident is compelled to spend the rest of his (her) life in a hemiplegic state. On the contrary, it is known that a peripheral nervous system (PNS)-neuron retains a vigorous regeneration ability even in those higher organisms, and therefore, neurons in a limb, when disconnected, can gradually regenerate with a concomitant recovery of their function.
In early nineteen-eighties, a group of Aguayo et al. found that when PNS-neuron is experimentally grafted into an injured CNS-neuron in a higher organism, axon growth of CNS-neuron is induced. This observation demonstrates that CNS-neuron in higher organisms which had been generally considered not to have a regeneration ability can regenerate if a suitable environment is provided (Nature, 284, 264-265 (1980), Science, 214, 931-933 (1981)). That report suggests a possibility that in CNS of higher organisms, there may exist a factor, namable “CNS-neuron regeneration inhibitor”, which inhibits the regeneration of CNS-neuron, and that a release from such inhibition may allow the regeneration of CNS-neurons. This suggestion paved the way for a CNS-neuron regeneration therapy.
In 1988, a group of Schwab et al. demonstrated that there existed such CNS-neuron regeneration inhibitor among proteins derived from CNS myelin. They also succeeded in purifying, though partially, a protein having said CNS-neuron regeneration inhibition activity, and named this protein fraction NI35/250 (Annu. Rev. Neurosci., 16, 565-595 (1993)), although no one has succeeded in its isolation, identification and gene cloning yet. In addition, they immunized animals with the partial purified NI35/250, and succeeded in obtaining an antibody (IN-1) having a neutralizing activity. This antibody is capable of recognizing a band for NI35/250 in Western blotting, and capable of staining, in an immunostaining, the region where NI35/250 is supposed to be distributed. Furthermore, they demonstrated that administration of this antibody to an animal experimentally received an injury on its spinal cord has promoted regeneration of axons in spinal cord, though partially, within 2-3 weeks, and restored its function within 2-3 months (Nature, 343, 269-272 (1990), Nature, 378, 498-501 (1995)). These findings are of great value, because they experimentally demonstrated that there existed a CNS-neuron regeneration inhibitor as suggested by Aguayo et al. (supra) and that CNS-neuron can be regenerated by inhibiting the activity of said inhibitor. The above noted antibody is, however, directed not to human but to rat NI35/250, and exhibits a low stability and specificity. In addition, although regeneration of CNS-neuron was observed as described above by administering said antibody, its effect was so partial and incomplete that not all of the motor functions could be restored. It is, therefore, believed essential in solving these problems to identify the gene coding for NI35/250 or corresponding CNS-neuron regeneration inhibitor, and, based on knowledges of molecular biology, neuroscience and the like, develop an inhibitor effectively inhibiting the CNS-neuron regeneration inhibition activity, or develop a method for inhibiting the expression of the gene for said regeneration inhibitor.
Apart from the above, the nervous system, whether it is central or peripheral, requires formation of a complicated neural network among neurons or between neurons and peripheral receivers or effectors during development, that is, in the stage of embryo or fetus, in order to precisely carry out its principal functions, i.e., to transfer and process the information. To establish the neural network, an ingenious mechanism is necessary, which precisely guides a growing neurite to the target site locating remote therefrom.
It has been hitherto believed that a factor which positively control the neurite outgrowth such as neurite growth promoter and neurite growth attractant may play a major role in the formation of the neural network. However, it is now being demonstrated by recent studies on the mechanism of the network formation that the opposite factor, that is, a negative factor having an outgrowth inhibition activity is important for an accurate guidance (Cell, 78, 353-356 (1994)).
A representative factor having such an outgrowth inhibition activity is a protein called “Semaphorin”. Semaphorin firstly discovered is Fasciclin IV found in grasshopper. Collapsin (latterly named Collapsin I) was subsequently discovered in chick (Cell, 75, 217-227 (1993); Neuron, 9, 831-845 (1992)). To date, more than 10 genes belonging to the Semaphorin family have been reported in a wide range of species covering insects such as drosophila and beetle, human, and viruses (Cell, 81, 471-474 (1995)). These Semaphorins characteristically contains in their amino acid sequences a certain structure called semaphorin domain consisting of about 500 amino acids (Neuron, 14, 941-948 (1995); Cell, 75, 1389-1399 (1993)). However, the homologies of the primary amino acid sequences in Semaphorin domains among these Semaphorin genes are 80-20%, and not necessarily high.
Of these Semaphorins, functions have been verified for only a few, including, for example, Fasciclin IV of grasshopper, Semaphorins I and II of drosophila, Collapsin of chick, and Semaphorin III which corresponds to Collapsin in mammals. All of these Semaphorins are known to inhibit neurite outgrowth and synapsis formation. In particular, Semaphorin III has been reported to have an activity collapsing in a short time the growth cone of cultured neuron (growth-cone collapse activity) in vitro (Neuron, 14, 941-948 (1995); Neuron, 14, 949-959 (1995); Cell, 81, 631-639 (1995); Cell, 75, 1389-1399 (1993); Cell, 75, 217-227 (1993); Neuron, 9, 831-845 (1992)).
Although it is now being demonstrated, as described above, that Semaphorin has a growth-cone collapse activity and a neurite outgrowth inhibition activity during development, and plays a role in giving an accurate guidance to neuron, it is not evident at present whether or not Semaphorin exerts some function not only during development but also in the adult, and less evident whether or not Semaphorin plays a role as a CNS-neuron regeneration inhibitor. Of course, since Semaphorin has been shown to be a negative guidance factor inhibiting neurite outgrowth, it would not be unreasonable to consider said Semaphorin as a candidate for a CNS-neuron regeneration inhibitor (Nature, 378, 439-440 (1995)). However, it has been shown by in vitro experiments that Semaphorin III (Sema III), only one Semaphorin of higher organisms of which function has been analyzed, exerts its neurite-outgrowth inhibition activity on a sensory neuron and sympathetic neuron both of which are peripheral, but not on a retinal neuron which is central (Cell, 75, 217-227 (1993)). In addition, Northern analysis on the distribution of Sema III expression in the adult conducted by the present inventors has revealed that it is expressed mainly in peripheral tissues (see Reference example 2 below). It is therefore hardly believed that Sema III having such features has a function as a “CNS-neuron regeneration inhibitor”.